Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks28 from public data
- A Sanity Check for AI-generated Image Detection256
A sanity check on whether the task of AI-generated image detection has been solved is conducted, and AIDE (AI-generated Image DEtector with Hybrid Features), which leverages multiple experts to simultaneously extract visual artifacts and noise patterns is proposed.
- WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMs170
The first benchmark to assess the multi-modal video understanding, that simultaneously encompasses visual, audio, and text inputs, is introduced and the experimental results indicate that existing models face significant challenges in understanding real-world scenarios.
- LamRA: Large Multimodal Model as Your Advanced Retrieval Assistant129
This paper introduces LamRA, a versatile framework designed to empower LMMs with sophisticated retrieval and reranking capabilities, and adopts a two-stage training strategy comprising language-only pre-training and multimodal instruction tuning to progressively enhance LMM’s retrieval performance.
- DynaPrompt: Dynamic Test-Time Prompt Tuning48
DynaPrompt is proposed, short for dynamic test-time prompt tuning, exploiting relevant data distribution information while reducing error accumulation, and a dynamic prompt selection strategy based on two metrics: prediction entropy and probability difference is introduced.
- 29
- Feature Augmented Memory with Global Attention Network for VideoQA29
A novel videoQA framework which progressively refines the representations of videos and questions from fine to coarse grain in a sequence-sensitive manner and achieves state-of-the-art performance on VideoQA benchmark datasets.
- DeViT22
Experimental results demonstrate that the Deformed Vision Transformers architecture outperforms previous state-of-the-art methods in quality and quantity and achieves a new state- of- the-art for video inpainting.
- Object-Centric Video Question Answering with Visual Grounding and Referring19
This paper introduces a VideoLLM model, capable of performing both object referring for input and grounding for output in video reasoning tasks, and proposes STOM (Spatial-Temporal Overlay Module), a novel approach that propagates arbitrary visual prompts input to the remaining frames within a video.
- 15
- Association of age at menarche and menopause, reproductive lifespan, and stroke among Chinese women: results from a national cohort study6
Subgroup analysis showed that association between age at menarche, reproductive lifespan and stroke events showed Chinese women with late age at menarche and shorter reproductive lifespan have higher risk of stroke.
- Deep Interactive Video Inpainting: An Invisibility Cloak for Harry Potter6
This is the first deep learning-based interactive video inpainting framework that only uses a free form of user input as guidance (i.e. scribbles) instead of mask annotations, which has academic, entertainment, and commercial value.
- A Multi-Group-Supporting Policy Hidden Fine-Grained Data Sharing Scheme in 5G-Enabled IoT With Edge Computing5
This work exploits a group key management scheme based on the Chinese Remainder Theorem to achieve efficient dynamic group management in IoT and implement concealment of data owners’ access policies to enhance privacy preservation.
- Preference-Aware Rubric Learning for Personalized Evaluation3
Experiments on real-world personalized text generation tasks show that PARL consistently induces high-fidelity rubrics that reliably identify user-aligned responses and generalize across users and tasks, while capturing stable stylistic preferences and fine-grained evaluative patterns.
- A comprehensive and efficient CP-ABE scheme with full policy hiding and pre-decryption verification in peer-to-peer cyber-physical systems3
This work proposes a fully policy-hidden CP-ABE scheme based on Linear Secret Sharing Scheme (LSSS) access structure that satisfies the security requirements for data sharing in CPS while exhibiting good scalability.
- Weaver: End-to-End Agentic System Training for Video Interleaved Reasoning2
Weaver is a novel, end-to-end trainable multimodal reasoning agentic system that empowers its policy model to dynamically invoke diverse tools throughout the reasoning process, enabling progressive acquisition of crucial visual cues and construction of authentic multimodal reasoning trajectories.
- VideoSeeker: Incentivizing Instance-level Video Understanding via Native Agentic Tool Invocation2
VideoSeeker seamlessly integrates agentic reasoning with instance-level video understanding tasks, enabling the model to proactively perceive and retrieve relevant video segments on demand, and constructs a four-stage fully automated data synthesis pipeline to efficiently generate large-scale, high-quality instance-level video data.
- Conditional Kronecker Batch Normalization for Compositional Reasoning.2
A novel CBN method using the Kronecker transformation, termed as Conditional Kronecker Batch Normalization (CKBN), which outperforms previous CBN methods in compositional reasoning, counting as well as the convergence speed on CLEVR dataset.
- AgentCVR: Active Multi-Agent Cross-Video Reasoning via Script-Simulated Reinforcement Learning1
AgentCVR is proposed, a multi-agent framework that treats CVR as an active evidence-acquisition task and achieves comparable performance to state-of-the-art closed-source systems, particularly in complex cross-video alignment and localization.
- Efficient and Expressive Public Key Authenticated Encryption with Keyword Search in Multi-user Scenarios1
This work proposes an efficient and expressive PAEKS scheme that obfuscates the mapping relationship associated with the LSSS matrix to the keywords, thereby enhancing the privacy protection and practicality and efficiency in multi-user data sharing scenarios.
- 1
- –
- –
- –
- –
- –
- –
- –
- EFFECT OF A WEB-BASED MANAGEMENT PLATFORM ON HYPERTENSION CONTROL IN COMMUNITY: A RANDOMIZED CLINICAL TRIAL–
A cluster andomised clinical trial of a hypertension management program conducted in 66 communities found that a multi-component intervention delivered by a web-based platform appeared to be more effective than usual care, leading to measurable benefits such as lower blood pressure, improved hypertension control, and adoption of healthy lifestyle habits.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.